Microbial Air Sampling in Pharmaceutical Aseptic Monitoring

Regulatory maximum colony forming units (CFUs) for a particular class or grade of cleanroom depends on; (1) the class or grade of the cleanroom; and (2) the regulatory agency (FDA, European Medicines Agency, etc.). According to WHO (World Health Organization), “The minor numerical differences that exist between EU, WHO, and ISO limits are not considered statistically significant.” Subsequently, a majority of pharmaceutical producers use a conflation of the most conservative elements of these standards. In aseptic processing, also known as critical areas, this is usually an average of less than 1 cfu per cubic meter when sampled under dynamic conditions (in operation).

An often-asked question is, “What is the best practice regarding the frequency of particulate and viable aseptic monitoring?” Recommendations are provided by WHO and USP <1116>. Monitoring frequency should be made in accordance with the manufacturer’s risk assessment.

Table 1: Recommended Monitoring Frequencies for “in operation” Particulate Monitoring.

Grade A (filling operation)	Full duration of the operation
Grade B	Daily*
Grade C	Weekly
Grade D	Not Required – periodic validation is sufficient
UDAF* workstations in B	Daily*
UDAF workstations in C	Weekly
UDAF workstations in D	Monthly
UDAF in UNC areas	Routing re-qualification of UDAF is sufficient

* UDAF: Uni-Directional Airflow

Table 2: Recommended Monitoring Frequencies for “in operation” Viable Monitoring.

Grade A (filling operation)	Full duration of the operation
Grade B	Daily
Grade C	Weekly
Grade D	Not Required
UDAF workstations in B	Daily
UDAF workstations in C	Weekly
UDAF workstations in D	Monthly
UDAF in UNC areas	Routing re-qualification of UDAF is sufficient

 

In biopharmaceutical manufacturing there are several criteria that one should consider when engaged in aseptic processing.

The FDA and GMP recommend a review of various processes periodically to ensure continual improvement. If it has been several years since the microbial sampler URS and SOP were reviewed, now might be a good time to re-assess your equipment requirements.

Unlike particle counters, microbial samplers have no regulatory minimum standards for performance. This is somewhat disconcerting for many who liken the regulatory standards for microbial sampling as similar to where particle counters were two to three decades ago. Others have referred to biopharma requirements for microbial samplers as similar to the “Wild West.”

Since the mid-1990s, various studies have demonstrated several facts regarding viable monitoring, which today we know to be true:

1. There is no such thing as 100 percent collection efficiency. As physical collection efficiency approaches 100 percent, impaction velocities cause a substantial decline in the percentage of microorganisms recovered.⁴
2. Viable microorganisms are stressed both through aerosolization as well as impaction causing further reduction in collection efficiency.⁴
3. The smaller the colony forming unit (CFU), the more difficult these are to collect as they tend to become entrained in the sampler’s air flow, and miss impacting on the agar.
4. Actual collection efficiency between samplers both in the field and the laboratory can have considerable differences in collection performance.^{4, 7}
5. Airborne viable microorganisms are not single cells, or free floating, but are commonly found in the form of microbe-carry-particles sized between 10 µm to 20 µm.⁶

Collection efficiency of microbial air samplers can be considered in two ways: Physical efficiency and biological efficiency. Physical efficiency is the ability to collect various sizes of particles, regardless if the particle is inert or viable. Biological efficiency is the ability to collect aerosolized microbe-carrying particles without destroying or mechanically damaging the microorganism, or causing the breakup of colony forming units. Biological efficiency will be lower than physical efficiency for a number of reasons.

ISO 14698 also states that, “A sampler must have good performance down to 1 µm.”¹ In the Life Science industry, this has generally been accepted to be at or near 50 percent physical recovery at 1 µm (d50 = 1µm). Unfortunately, very few microbial air samplers today actually meet this requirement, and in fact most impaction samplers well exceed 5 µm.⁸ Since we know aerosolized microbe-carrying particles are 10 µm to 20 µm in size, why is good performance down to 1 µm important? The answer is simple, small particles are more difficult to collect than larger macroparticles (> 5µm). Therefore the physical collection of particles is, from a graphical perspective, upwards sloping. More simply, the lower the d50 value, the higher the sensitivity and accuracy when macroparticles exceed 5 µm.  

When a biopharmaceutical manufacturer is evaluating Operational Efficiency of a microbial sampler, we must remember that the term Collection Efficiency is in an informative (not normative) section of ISO 14698, and the test has nothing to do with actual efficiency, but rather it is a comparison equation:

Efficiency of Sampler (%) =           Test Sampler Count           x 100
                                                    Reference Sampler Count

According to ISO 14698, at least 10 experiments should be taken with the sampler under test side-by-side with the reference sampler.

The looseness of this standard and inability of most independent laboratories to even perform the testing described has also created an environment where it is possible a manufacturer may manipulate the testing to optimize their test results and ultimately positively affect their product’s marketability. For example, substantially different results can be achieved with the same microbial sampler when compared to, for example, the industry gold standard in biological efficiency vs. a low efficiency sampler.

Subsequently, it is recommended biopharma managers ask the manufacturer of the microbial sampler the following questions:

1. What are your physical and biological collection efficiencies?
2. What is your microbial sampler’s d50 value?
3. What is the reference sampler’s d50 value?
4. Please, provide a copy of your biological efficiency test report.
   1. This report should be written and compiled from an independent laboratory.
   2. This report will confirm the methodology, test parameters, as well as the manufacturer and model of the reference sampler.
   3. The report will display the ten (or more) experiments taken from the sampler under test and the reference sampler.
5. What is the reason you chose this make and model as the reference sampler?
6. Why did you test with this particular strain of bacteria? (the species will be mentioned in the laboratory test report)
7. Repeatability of measurement: Metrologists will confirm that the two most important factors in any measurement instrument is accuracy and repeatability. Once again, there are no standards to help users determine measurement repeatability as it pertains to a microbial sampler. However, if you have the independent laboratory test report, this will list the results of each comparative sample for the sampler under test, as well as the reference sampler. Quality Managers should calculate the correlation coefficient (r) for these two datasets. Ideally, the sampler under test and the reference sampler should have a positive near perfect linear relationship (r > +0.95). This will confirm if there is strong repeatability between the two instruments.

These answers when compared against the various commercially available models should be able to help glean a much better understanding of which models truly have superior operational performance.

Another factor unique to aseptic processing is overall quality. In biopharmaceutical manufacturing the major cost driver for microbial samplers is not the initial purchase price, but rather the cost of poor quality. Microbial samplers require calibration at least annually, and some drug manufacturers, particularly those associated with parenteral drug solutions, frequently perform calibration immediately before and after each production run. When the microbial sampler is found to be Out-of-Tolerance (OOT) during its interval calibration, producers in regulated industries are required to conduct a deviation investigation in order to ensure the quality of the product produced is not compromised.

The cost for conducting this investigation is a hidden cost consisting of almost all labor, involving multiple departments, which includes manufacturing burden, etc. The average cost for a simple failure investigation among pharmaceutical producers averages of $8,000 to $12,000 per incident. The cost of a single OOT will often well exceed the price of a microbial sampler.

Consequently, biopharma producers should pay close attention to the OOT rates of a microbial sampler. Per the below chart, if we apply the law of continuous probabilities:

OOT Rate	Probability of passing 5 consecutive interval calibrations
1%	95.1%
6%	73.4%
15%	44.4%

HEPA filtered exhaust is of paramount importance in parenteral drug manufacturing, where one of the top causes for batch rejection and product recall are foreign particulate contamination.⁵ Unlike biocontamintion, which is heavily tied to cleanroom personnel, foreign particulate contamination comes almost exclusively from production and laboratory equipment and generally consists of glass, silicone, plastic, and stainless steel.⁶ These materials, when injected into an unhealthy patient, may prove fatal.

As a consequence, parenteral drug manufacturers will spend millions maintaining and testing cleanroom HEPA filters, HEPA filters on biological safety cabinets, and filtering on other exhaust emissions. These same manufacturers will insist particle counters used in aseptic routine monitoring have HEPA filtered exhaust. However, an estimated 80 to 90 percent of all parenteral drug manufacturers use microbial samplers that have no HEPA filtered exhaust!

When a microbial sampler or particle counter has a leaky HEPA filtered exhaust, this can cause what the industry has termed “rogue emissions.” These rogue emissions (or inert particles) are caused by moving parts such as blowers and fans, i.e. vacuum sources. A leaky HEPA filter will emit one to two thousand particles each time a cubic meter is sampled. A microbial sampler without a HEPA filter will exhaust from the tens of thousands to hundreds of thousands of inert particles each time a cubic meter sampled; 97 percent of these particles fall within the >0.5 µm range, and are small enough to become aerosolized and entrained in turbulent air spreading across an entire production area.

Other factors which are important considerations in pharmaceutical aseptic processing include:

1. Stainless steel enclosure. One of the few static neutral materials, stainless steel does not attract airborne particles or microbe-carrying-particles. It is the easiest material to clean and disinfect. At the opposite end of the spectrum are plastics, which can carry a high negative static charge and will attract particles of every size, including aerosolized microorganisms. In addition to attracting microorganisms, plastics are actually a food source for some bacteria (Ideonella sakainesis). Finally, plastics will biodegrade when exposed to UV light, moisture, exposure to harsh chemicals, oxygen, heat, or enzymes excreted by bacteria. During the biodegradation process, micro-fracturing will occur, and inert particles will also be released into the clean zone. Moreover, the cracking and micro-fractures are difficult to disinfect and sanitize.
2. Autoclavable sample heads: The depth of each hole on a sample head is about 3000 µm, and make perfect hiding places for contamination.
3. Remote sampling capability in biological safety cabinets, isolators, or RABS.²
4. Ability to sample High Pressure Gases.⁶

 

Quite simply, biopharmaceutical manufacturers need to ensure their choice of environmental monitoring equipment used in aseptic processes truly has the features, sensitivity, accuracy, and repeatability to monitor those critical areas that pose substantial risk not only to their organization’s balance sheet, but more importantly to patient health.

References

1. ISO 14698-1:2003
2. Center for Disease Control. Biosafety in Microbiological and Biomedical Laboratories, 5th Edition. HHS Publication No. (CDC) 21-1112. U.S. Department of Health and Human Services December 2009.
3. Lin, Reponen, Willeke, Wang, Grinshpun, and Trunov. Survival of Airborne Microorganisms During Swirling Aerosol Collection. Aerosol Science and Technology, 32:184-196 (2000). American Association of Aerosol Research.
4. Stewart, Grinshpun, Terzieva, Ulevicius and Donnelly. Effect of Impact Stress on Microbial Recovery on an Agar Surface. Applied and Environmental Microbiology, Vol. 61, No. 4, Apr 1995, p. 1232-1239. American Society for Microbiology.
5. Tawde SA (2015) Particulate Matter in Injectables: Main cause for Recalls. J Pharmacovigil 3: e128. doi: 10.4172/2329-6887.1000e128
6. USP <1116>, Microbial Control and Monitoring of Aseptic Processing Environments. August 1, 2013.
7. Whyte, Green, and Albisu. Collection Efficiency and Design of Microbial Air Samplers. Department of Mechanical Engineering, University of Glasgow, Scotland. 20 May 2014.
8. Yao, Mainelis. Investigation of Cut Off Sizes and Collection Efficiencies of Portable Microbial Samplers. Aerosol Science and Technology, 40:595-606 (2006). American Association of Aerosol Research.

Jim Strachan is General Manager of Climet Instruments Co. in Redlands, Calif. [email protected]; www.climet.com